Rate a few songs to see how your taste in music compares with people around the world!
Your ratings also help with our research on better music recommendations.
This study requires that your browser be able to play sound. Play test sound.About
You're about to participate in an online experiment. Your contribution to our research helps us learn about qualities and attributes of music. The results from your test will also tell you something about yourself!
Please read the following information carefully before continuing.
Why we are doing this research: We are trying to understand how music preferences relate to identifiable attributes about how it sounds.
What you will have to do: You will listen to a set of songs and we will ask you to rate them based on various attributes. Refusal to participate will involve no penalty or loss of benefits to which you are otherwise entitled.
Potential risks: There are no risks anticipated in taking part in this study and you are free to leave at any time without penalty or loss of benefits to which you are otherwise entitled.
Length of Test: Approximately 10 minutes.
To contact the researcher: If you have questions or concerns about this research, please contact Kenneth Arnold, Maxwell Dworkin 217, 33 Oxford St, Cambridge, MA 02138, [email protected] You may also contact the faculty member supervising this work: Prof. Krzysztof Gajos, 33 Oxford St, Cambridge, MA 02138, [email protected]
Whom to contact about your rights in this research, for questions, concerns, suggestions, or complaints that are not being addressed by the researcher, or research-related harm: Committee on the Use of Human Subjects in Research at Harvard University, 1414 Massachusetts Avenue, Second Floor, Cambridge, MA 02138. Phone: 617-496-2847 (CUHS). Email: [email protected]You must check the box to continue. About
You've rated # songs. You can stop whenever, but it's best if you could rate at least 20.
We picked random songs from the Million Song Dataset, filtered to be from familiar artists. We suspect that classical music is under-represented because though composers may be familiar, the artists who actually recorded them might not be.
People generally listen to different music in different situations. We're working on an app that accounts for situations in its recommendations. To help it learn faster, we're letting people say what attributes of the song make it appropriate or not for that situation. Here, you're helping us learn how those attributes apply to other music.
You can also participate in the companion experiment to provide examples of those particular music-listening scenarios.
For other experiments where you find out about yourself while also contributing to scientific research, check out Lab in the Wild!
I'm Ken Arnold at Harvard IIS. You can email me with any questions or comments at [email protected].